To compare the effects of electroacupuncture (EA) with different time intervals on corticospinal excitability of the primary motor cortex (M1) and the upper limb motor function in healthy subjects and observe the after-effect rule of acupuncture. Self-comparison before and after intervention design was adopted. Fifteen healthy subjects were included and all of them received three stages of trial observation, namely EA0 group (received one session of EA), EA6h group (received two sessions of EA within 1 day, with an interval of 6 h) and EA48h group (received two sessions of EA within 3 days, with an interval of 48 h). The washout period among stages was 1 week. In each group, the needles were inserted perpendicularly at Hegu (LI 4) on the left side, 23 mm in depth and at a non-acupoint, 0.5 cm nearby to the left side of Hegu (LI 4), separately. Han's acupoint nerve stimulator (HANS-200A) was attached to these two needles, with continuous wave and the frequency of 2 Hz. The stimulation intensity was exerted higher than the exercise threshold (local muscle twitching was visible, and pain was tolerable by healthy subjects, 1-2 mA ). The needles were retained for 30 min. Using the single pulse mode of transcranial magnetic stimulation (TMS) technique, before the first session of EA (T0) and at the moment (T1), in 2 h (T2) and 24 h (T3) after the end of the last session of EA, on the left first dorsal interosseous muscle, the amplitude, latency (LAT), resting motor threshold (rMT) of motor evoked potentials (MEPs) and the completion time of grooved pegboard test (GPT) were detected. Besides, in the EA6h group, TMS was adopted to detect the excitability of M1 (amplitude, LAT and rMT of MEPs) before the last session of EA (T0*). The amplitude of MEPs at T1 and T2 in the EA0 group, at T0* in the EA6h group and at T1, T2 and T3 in the EA48h group was higher when compared with the value at T0 in each group separately (P<0.001). At T1, the amplitude of MEPs in the EA0 group and the EA48h group was higher than that in the EA6h group (P<0.001, P<0.01); at T2, it was higher in the EA0 group when compared with that in the EA6h group (P<0.01); at T3, the amplitude in the EA0 group and the EA6h group was lower than that of the EA48h group (P<0.001). The LAT at T1 was shorter than that at T0 in the three groups (P<0.05), and the changes were not obvious at the rest time points compared with that at T0 (P > 0.05). The GPT completion time of healthy subjects in the EA0 group and the EA48h group at T1, T2 and T3 was reduced in comparison with that at T0 (P<0.001). The completion time at T3 was shorter than that at T0 in the EA6h group (P<0.05); at T2, it was reduced in the EA48h group when compared with that of the EA6h group (P<0.05). There were no significant differences in rMT among the three groups and within each group (P>0.05). Under physiological conditions, EA has obvious after-effect on corticospinal excitability and upper limb motor function. The short-term interval protocol (6 h) blocks the after-effect of EA to a certain extent, while the long-term interval protocol (48 h) prolongs the after-effect of EA.